and night-time illumination probably made the fauna of the near side 

 of this reef behave in an abnormal fashion, hence we concentrated on 

 the opposite side of the reef from the habitat. However, we supple- 

 mented our normal morning and evening rebreather dives with observa- 

 tions from the portholes and by short excursions on hookah, especially 

 when adverse weather conditions limited us to the habitat (due to 

 restrictions on surface support). "Sewer Reef," so named because the 

 easiest route to the reef was along the sewer line from the habitat, 

 was chosen for its accessibility (a 5-10 minute swim from the habitat), 

 lack of human interference, and greater depth (23 m) . We picked a 

 deeper reef as the second study site because it was easily available 

 to us and more difficult of access by surface-based workers. Also, 

 two other members of Mission 15-50 (C. L. Smith and J. C. Tyler) were 

 intensively working a shallower (11 m) reef. 



Morning and evening observations were conducted, beginning before any 

 changes in activicy patterns and continuing until all major changes 

 had occurred. This plan entailed dives approximately two hours long 

 at about 0500-0700 and 1700-1900 hours eastern daylight time. Notes 

 were recorded by both of us at 5-10 minute intervals on reef fish 

 activities--resting, feeding, hovering in the water column, parasite 

 picking, etc. Observations were concentrated on basic activity 

 patterns--when a given species was active and when it was "sleeping" 

 or resting. We also recorded day-night color pattern differences. 

 Notes were transcribed from the underwater slates after each dive or 

 two and then summarized by species. 



RESULTS 



Morning and evening changeovers were defined based on the time of 

 replacement of nocturnal by diurnal species in the morning and the 

 reverse process in the evening. We identified 107 species of fishes 

 in 36 families; data is of reasonable quality for about 35 species. A 

 species by species account summarizing our observations has been 

 prepared for publication elsewhere. 



The first appearance of diurnal species was as early as 0540 hours and 

 all were usually active before 0645 hours. Species in some groups of 

 fishes, for example the damselfishes (Pomacentridae) were early risers 

 (mean times of first appearance 0556-0605 hours) and remained active 

 throughout the day (until 1743-1808 hours). Species in other groups, 

 such as the parrotfishes (Scaridae) and wrasses (Labridae) became 

 active later (mean times of first appearance 0613-0618 hours) and dis- 

 appeared sooner (1735-1756 hours). The sparse nocturnal fish fauna 

 (primarily cardinalf ishes and squirrelfishes) became active between 

 1812 and 1835 hours, after most of the many colorful diurnal fishes 

 had already disappeared (1732-1812 hours). The best example of the 

 replacement of one fauna by another involves midwater particulate 

 plankton feeders. The cardinalfish Apogon affinis moved out of its 

 daytime caves at about 1800 hours to replace the diurnal aggregations 

 of the damselfishes Chromis cyanea and C. multilineata . At morning 



VI-258 



